DESCRIPTION (Taken from application abstract): In order to examine the mechanisms of neurodegeneration in the human brain it is essential to employ methods that allow the analysis of individual neurons. This requires the use of histochemical methods where individual features of cells can be identified. Recently, it has become possible to use computer-based methods to obtain quantitative data on histochemical preparations that saves time and leads to more precise results. In this proposal our goal is to examine the progression of neuronal degeneration in the brains of patients who have died of Alzheimer's disease (AD) with the aid of a computer based cellular recognition system that analyzes cellular features similar to the human eye. Once captured these data will be links to an object-oriented database that stores, retrieves and aids in the analysis of the data. Specifically, we will build on the model of disease staging and progression proposed by Braak and Braak and develop machine vision methods to capture and quantify the development of tangles. Storage of the data, higher level automation, and declarative retrieval of information is performed by an object-oriented database. Sections of key brain regions will be stained with antibodies to tangles and B-amyloid using two color double-label immunocytochemical methods. These will be analyzed for color separation, edge detection and the pattern of the relevant cellular cytoarchitectonic features. In this manner the quantitative relationship of staging within a brain region and across proposed regions will be determined. This data will be used to examine the progression of other indices of neuronal degeneration. Specifically, we will examine the relationship between tangle formation and DNA damage and test the hypothesis that in some cases DNA damage can occur prior to or independent of tangle formation. We will also examine the hypothesis that single strand breaks occur prior to double strand breaks and are accompanied by an induction of repair/protective mechanisms such as BCL-2. In this way we will be able to gain new information on the mechanisms leading to neuronal death in the brain. Importantly the basic method and approach will apply to many different markers and to other applications.